A sequential classification algorithm for autoregressive processes and its application on radar targets

A sequential method for the classification of the autoregressive processes is presented. Different from the conventional detectors having fixed sample size, the suggested method uses Wald´s sequential probability ratio test (SPRT) which is known with faster decisions and has a variable sample size. It is assumed that the coefficients of the filters generating the autoregressive processes are known; but the variance of process is unknown to the detector. Under these conditions, the generalized likelihood ratio (GLRT) is derived and its sequential operation described. Finally, the suggested algorithm is applied to the classification of fixed and rotary wing targets.